Abstract
Remotely sensed data is the most useful tool for observing vegetation globally and temporally. It is needed to estimate vegetation amount quantitatively from satellite data by building the indirect model using spectral information and biomass data measured in the field. It is necessary to develop the method to process the field data which is satisfied with condition of the simultaneity and spatial scale for satellite data. The aim of this study is to develop and verify the procedure using field data obtained by moving observation we developed in order to measure the quantitative vegetation amount globally in grassland using satellite data. The conducted study area is Mongolian grassland which is very flat and homogeneous. By using moving observation system, we could obtain so many video images and spectral reflectance at the same time. We analyzed about 300 spectral information, about 300 video images and about 10 biomass data by applying RGB Pattern method we proposed for estimating relationship between vegetation coverage ratio and 4 vegetation indices (NDVI, SAVI (0.5), MSAVI, APVI) . We calculated these vegetation indices considering NOAA AVHRR sensor Ch. 1 and Ch. 2. Finally, we verified relationship between vegetation coverage ratio and vegetation indices, which is estimated by linear regression analysis. The summary of results is shown below
1. we could obtain many field data at the satellite scale.
2. we developed the method which calculate VCR automatically from many video images together.
